Human chromosome 5 is estimated to be approximately 150Mb in length and to encode 5,000 genes. We propose embarking on a physical map of the majority of these transcription units using chromosome 5-specific cDNAs that have been isolated by direct cDNA selection, in conjunction with a variety of physical mapping resources. A total of 50,000 chromosome 5-specific cDNAs from human fetal brain, placenta, activated lymphocytes, pooled mouse embryos and from a collection of normalized human infant brain cDNAs will be picked and arrayed into a high density format. Expressed Sequence Tags will be developed from 500 random, non redundant members of this array. These will be used to identify clones within an array of large insert YACs. Chromosome 5-specific YAC contigs, microdissection probes and radiation reduced hybrid DNAs will be converted into cosmids by hybridization to a chromosome 5-specific cosmid library. These cosmids will in turn, be used to identify cDNAs within the gridded cDNA array. Various normalized and primary cDNA sets will be hybridized en masse to the array to derive pattern and level of expression data on the arrayed cDNAs. Finally, oligonucleotide probes for several biologically interesting DNA sequence motifs will be used to screen the array to derive inferential data on potential functions for some members of the set. The final product of this work will be a reference set of arrayed cDNA clones, cosmids and YACs from chromosome 5 that will be linked together in an interrelated database. This database will also contain information on expression patterns for many of the genes on the chromosome. This should constitute an important resource for studying gene distribution, control and evolution, as well as assisting in the search for many of the genetic disease loci that are located on this chromosome.